Bayesian Analysis of First-Order Markov Models for Autocorrelated Binary Responses (original) (raw)
Journal of Statistical Theory and Practice
Related papers
2016
Bayesian approach has been increasingly applied to various aspects of design and analysis of clinical trials. We present one application concerning an interim futility analysis of a trial. Longitudinal data were collected for a range of the studied doses. Bayesian analysis was first conducted to predict observations at the end of treatment for patients not yet followed through treatment, based on all interim observed data. The predicted data in combination with observed data at the end of treatment were then analyzed using a Bayesian normal dynamic linear model for dose response inference. Summary of the Bayesian analysis was used to aid an interim futility decision.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.